Search: “reasoning”

DiBS: Diffusion-Informed Branch Selection

arXiv:2606.06518v1 Announce Type: new Abstract: Sudoku is a representative constraint satisfaction problem that requires global structural reasoning under strict discrete constraints. The existing wor

Neuro-Symbolic Learning for Long-Horizon Task Planning Under Complex Logical Constraints

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DGX agent

arXiv:2606.06877v1 Announce Type: cross Abstract: Task planning often suffers from severe efficiency bottlenecks when robots must reason over long-horizon action sequences under complex logical constr

safetyarxiv-cs-ai

8 Jun 2026

On the Geometry of On-Policy Distillation

arXiv:2606.07082v1 Announce Type: cross Abstract: On-policy distillation (OPD) is increasingly used to improve large language model reasoning, but its training dynamics remain poorly understood. We ch

TALAN: Task-Aligned Latent Adaptation Networks for Targeted Post-Training of Large Language Models

model-releasesarxiv-cs-lg

arXiv:2606.06902v1 Announce Type: new Abstract: Targeted post-training aims to improve reasoning, math, and code without degrading strengths. Low-rank adapters are efficient but task-global; activatio

Trio: Learning Time-Series Forecasting with Temporal-Spatial-Sample Attention and Structural Causal Priors

arXiv:2606.07291v1 Announce Type: new Abstract: Multivariate time-series forecasting requires models to reason over temporal dynamics, cross-variable dependencies, and historical input-output correspo

tutorialsarxiv-cs-lg

Where to Touch, How to Contact: Hierarchical RL-MPC Framework for Geometry-Aware Long-Horizon Dexterous Manipulation

arXiv:2601.10930v3 Announce Type: replace Abstract: A key challenge in contact-rich dexterous manipulation is the need to jointly reason over global geometry and nonsmooth contact dynamics. End-to-end

safetyarxiv-cs-ro

A Pre-Registered Causal Partition of Self-Consistency Elicitation and Reward Design in RLVR

arXiv:2606.05932v1 Announce Type: new Abstract: Reinforcement learning from verifiable rewards (RLVR) improves reasoning even when the reward signal is spurious -- assigning credit to the group-plural

Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads

arXiv:2606.06448v1 Announce Type: new Abstract: LLM agents are increasingly deployed on long-horizon tasks requiring sustained reasoning over extended interaction histories. Realizing this at scale re

Breaking the Chain: A Causal Analysis of LLM Faithfulness to Intermediate Structures

arXiv:2603.16475v2 Announce Type: replace Abstract: In schema-guided reasoning (SGR) pipelines, LLMs produce explicit intermediate structures -- rubrics, checklists, or verification queries -- before

researcharxiv-cs-ai

From Reward-Hack Activations to Agentic Risk States: Context-Calibrated Mechanistic Monitoring in LLM Agents

arXiv:2606.06223v1 Announce Type: new Abstract: Language-model agents act through repeated cycles of observation, reasoning, and action selection, making safety monitoring depend on both internal mode

Policy-Conditioned Counterfactual Credit for Verifiable Reinforcement Learning of Long-Horizon Language Agents

arXiv:2606.05263v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards improves reasoning and tool use, yet long-horizon language agents still learn unsupported evidence chai

When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents

arXiv:2606.05806v1 Announce Type: new Abstract: Existing benchmarks evaluate Tool-Integrated Reasoning (TIR) in LLMs on idealized ''happy paths'', largely overlooking real-world tool failures. We intr

Beyond Absolute Scores: Relative Edit-induced Difference for Generalizable Image Aesthetic Assessment

arXiv:2606.05778v1 Announce Type: new Abstract: Traditional Image Aesthetic Assessment (IAA) methods mainly rely on regressing absolute Mean Opinion Scores (MOS). However, such a paradigm overlooks th

researcharxiv-cs-cv

CLASH: Evaluating Language Models on Judging High-Stakes Dilemmas from Multiple Perspectives

model-releasesarxiv-cs-cl

arXiv:2504.10823v4 Announce Type: replace Abstract: Navigating dilemmas involving conflicting values is challenging even for humans in high-stakes domains, let alone for AI, yet prior work has been li

Cosine Misleads: Auxiliary Losses Reshape Vision Language Models, Not Their Latents

arXiv:2606.05753v1 Announce Type: new Abstract: Latent visual reasoning (LVR) inserts supervised latent tokens between perception and answer generation in vision-language models (VLMs). The field uses

safetyarxiv-cs-cv

Faithful, Enriched, and Precise: Benchmarking Natural-Science Illustration Generation by T2I models

model-releasesarxiv-cs-cv

arXiv:2606.05949v1 Announce Type: new Abstract: Scientific illustrations are essential tools for communicating research findings, especially in natural science, where they visualize complex concepts a

Formal Concept Lattices are Good Semantic Scaffolds for Concept-Based Learning

arXiv:2606.05471v1 Announce Type: new Abstract: Learning semantics is essential for deep learning models to be interpretable and better aligned with human reasoning. Concept-based models approach this

tutorialsarxiv-cs-cv

Global-Local Monte Carlo Tree Search in Vision-Language Models for Text-to-3D Indoor Scene Generation

arXiv:2606.06002v1 Announce Type: new Abstract: Large Vision-Language Models have achieved significant reasoning performance in various tasks.However, there are few studies on text-to-3D indoor scene

researcharxiv-cs-cv

Harnessing Generalist Agents for Contextualized Time Series

arXiv:2606.05404v1 Announce Type: cross Abstract: Time series are often embedded in rich contexts that are essential for holistic modeling. Moreover, real-world practitioners often require end-to-end

IR3DE: A Linear Router for Large Language Models

arXiv:2606.06098v1 Announce Type: new Abstract: Foundational Large Language Models (LLMs) demonstrate proficiency on a wide range of general tasks, and achieve remarkable results on various specialize

researcharxiv-cs-cl

Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation

arXiv:2606.06076v1 Announce Type: cross Abstract: While vision-language models excel at general multimodal understanding, they still struggle with visual spatial planning. We attribute this to a perce

safetyarxiv-cs-cv

On Advantage Estimates for Max@K Policy Gradients

arXiv:2606.06080v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards is widely used for post-training reasoning models, but sparse outcome rewards make exploration difficul

safetyarxiv-cs-cl

Tools

Quoting Andreas Kling

We will no longer accept public pull requests. [...] A substantial patch used to imply substantial effort, and that effort was a reasonable proxy for good faith. That assumption no longer holds. [...]

toolssimon-willison

Safe Embodied AI for Long-horizon Tasks: A Cross-layer Analysis of Robotic Manipulation

model-releasesarxiv-cs-ro

arXiv:2606.05660v1 Announce Type: new Abstract: Embodied AI systems are increasingly expected to reason and act over extended horizons in physical environments. This growing capability brings safety t

SkillComposer: Learning to Evolve Agent Skills for Specification and Generalization

arXiv:2606.06079v1 Announce Type: new Abstract: Agent skills, which consist of reusable strategies that guide agent reasoning and action, have shown strong potential for improving model capability at

The latest AI news we announced in May 2026

Google's May 2026 AI updates center on the new 'agentic' era, featuring the Gemini 3.5 model and Gemini Omni for advanced reasoning and creation. Gemini Omni is a new model that can create anything fr

model-releasesgoogle-ai

The Self-Correction Illusion: LLMs Correct Others but Not Themselves

arXiv:2606.05976v1 Announce Type: cross Abstract: Recent work shows that LLM agents struggle to correct errors in their own reasoning traces yet show markedly higher correction rates when identical cl

UNIVID: Unified Vision-Language Model for Video Moderation

arXiv:2606.05748v1 Announce Type: cross Abstract: Global-scale video moderation faces a dual challenge: the need for fine-grained multi-modal reasoning and the demand for interpretable outputs to supp

safetyarxiv-cs-cl

Unsupervised Skill Discovery for Agentic Data Analysis

arXiv:2606.06416v1 Announce Type: cross Abstract: Inference-time skill augmentation provides a lightweight way to improve data-analytic agents by injecting reusable procedural knowledge without updati

You Only Index Once: Cross-Layer Sparse Attention with Shared Routing

arXiv:2606.06467v1 Announce Type: new Abstract: Long-context inference in modern LLMs is increasingly constrained by decoding efficiency, especially in reasoning-heavy settings where models generate l

researcharxiv-cs-cl

Adaptive Minds: Empowering Agents with LoRA-as-Tools

arXiv:2510.15416v2 Announce Type: replace Abstract: We investigate a framework in which LoRA adapters are treated as callable tools that a base language model can dynamically select and invoke. We hyp

Agent Planning Benchmark: A Diagnostic Framework for Planning Capabilities in LLM Agents

model-releasesarxiv-cs-cl

arXiv:2606.04874v1 Announce Type: new Abstract: Planning is central to LLM agents: before acting, an agent must decompose goals, select tools, reason over constraints, and decide when a task is infeas

Arithmetic Pedagogy for Language Models

arXiv:2606.05106v1 Announce Type: cross Abstract: We investigate whether methods of human mathematics pedagogy can guide the training of language models toward arithmetic reasoning. Building on the GA

tutorialsarxiv-cs-ai

Hardware

Cartridges at Scale: Training Modular KV Caches over Large Document Collections

arXiv:2606.04557v1 Announce Type: new Abstract: Large Language Models can reason over long contexts, yet prefilling millions of tokens is wasteful as much of the content remains static across queries.

hardwarearxiv-cs-cl

Dynamic Infilling Anchors for Format-Constrained Generation in Diffusion Large Language Models

arXiv:2606.04535v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer bidirectional attention and parallel generation, enabling them to exploit global context and naturally s

researcharxiv-cs-ai

Emotion Entanglement and Bayesian Inference for Multi-Dimensional Emotion Understanding

arXiv:2604.00819v2 Announce Type: replace-cross Abstract: Understanding emotions in natural language is inherently a multi-dimensional reasoning problem, where multiple affective signals interact thro

Food-R1: A Unified Multi-Task Food Vision-Language Model with Reinforcement Learning

model-releasesarxiv-cs-cv

arXiv:2606.04986v1 Announce Type: new Abstract: Recent studies have explored Vision-Language Models (VLMs) for food analysis. However, most existing methods rely primarily on supervised fine-tuning (S

GeoMin: Data-Efficient Semi-Supervised RLVR via Geometric Distribution Modeling

arXiv:2606.04516v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) significantly advances LLM reasoning, yet it faces a dilemma: standard supervised scaling is thr

tutorialsarxiv-cs-ai

Hadamard thought in image space

Hadamard thought in image space World Labs CEO Dr. Fei-Fei Li: 'The world is not made of words.' 'Language models have given machines an extraordinary command of concepts, vocabulary, and reasoning, b

tutorialselon-musk--x

Rollout-Level Advantage-Prioritized Experience Replay for GRPO

arXiv:2606.04560v1 Announce Type: cross Abstract: Reinforcement learning from verifiable rewards with GRPO is a standard approach for post-training reasoning LLMs. It remains sample inefficient. Each

This is too good to keep to myself. I will be sharing more on how I am using dynamic workflows and how and why I built my own version of it:…

The post announces that the author will share insights about their use of dynamic workflows and explain their reasoning behind building a custom implementation. This likely covers practical applicatio

researchdair-ai--x

UniFine: A Unified and Fine-grained Approach for Zero-shot Vision-Language Understanding

arXiv:2307.00862v3 Announce Type: replace-cross Abstract: Vision-language tasks, such as VQA, SNLI-VE, and VCR are challenging because they require the model's reasoning ability to understand the sema

researcharxiv-cs-cl

⚠️Why didn’t the hyperscalers wait until after the IPO in switching to pay-by-usage-charging? My guess is that *they literally could not aff…

⚠️Why didn’t the hyperscalers wait until after the IPO in switching to pay-by-usage-charging? My guess is that *they literally could not afford to* —because it would bankrupt them. My reasoning: in “a

agentsgary-marcus--x

A workflow audit is no longer the best way to figure out how to use AI in your job. Despite the advice from AI labs, I'm more convinced, bec…

model-releasesallie-k--miller--x

A workflow audit is no longer the best way to figure out how to use AI in your job. Despite the advice from AI labs, I'm more convinced, because of AI's reasoning capabilities and long context horizon

Tools

[AINews] Microsoft Build: MAI-Thinking-1 and MAI Family models

Microsoft announced the MAI-Thinking-1 and MAI family of models at their Build conference, representing new advances in their AI model lineup. These models likely focus on improved reasoning capabilit

toolslatent-space

Align-KD: Distilling Cross-Modal Alignment Knowledge for Mobile Vision-Language Model Enhancement

arXiv:2412.01282v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) bring powerful understanding and reasoning capabilities to multimodal tasks. Meanwhile, the great need for capab

Breaking the Self-Confirming Loop: Diagnosing and Mitigating Systemic Reward Bias in Self-Rewarding RL

arXiv:2510.08977v2 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards (RLVR) efficiently scales the reasoning ability of large language models (LLMs) but is bottlene

safetyarxiv-cs-cl

Evaluating LLMs' Effectiveness on Real-World Consumer Device Repair Questions

arXiv:2606.03331v1 Announce Type: cross Abstract: Consumer device repair is an important but underexplored testbed for large language models (LLMs). Repair tasks require reasoning over incomplete prob